VALD: Multi-Stage Vision Attack Detection for Efficient LVLM Defense
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917349475483648 |
|---|---|
| author | Kadvil, Nadav Fares, Malak Tal, Ayellet |
| author_facet | Kadvil, Nadav Fares, Malak Tal, Ayellet |
| contents | Large Vision-Language Models (LVLMs) can be vulnerable to adversarial images that subtly bias their outputs toward plausible yet incorrect responses. We introduce a general, efficient, and training-free defense that combines image transformations with agentic data consolidation to recover correct model behavior. A key component of our approach is a two-stage detection mechanism that quickly filters out the majority of clean inputs. We first assess image consistency under content-preserving transformations at negligible computational cost. For more challenging cases, we examine discrepancies in a text-embedding space. Only when necessary do we invoke a powerful LLM to resolve attack-induced divergences. A key idea is to consolidate multiple responses, leveraging both their similarities and their differences. We show that our method achieves state-of-the-art accuracy while maintaining notable efficiency: most clean images skip costly processing, and even in the presence of numerous adversarial examples, the overhead remains minimal. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_19570 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | VALD: Multi-Stage Vision Attack Detection for Efficient LVLM Defense Kadvil, Nadav Fares, Malak Tal, Ayellet Computer Vision and Pattern Recognition Large Vision-Language Models (LVLMs) can be vulnerable to adversarial images that subtly bias their outputs toward plausible yet incorrect responses. We introduce a general, efficient, and training-free defense that combines image transformations with agentic data consolidation to recover correct model behavior. A key component of our approach is a two-stage detection mechanism that quickly filters out the majority of clean inputs. We first assess image consistency under content-preserving transformations at negligible computational cost. For more challenging cases, we examine discrepancies in a text-embedding space. Only when necessary do we invoke a powerful LLM to resolve attack-induced divergences. A key idea is to consolidate multiple responses, leveraging both their similarities and their differences. We show that our method achieves state-of-the-art accuracy while maintaining notable efficiency: most clean images skip costly processing, and even in the presence of numerous adversarial examples, the overhead remains minimal. |
| title | VALD: Multi-Stage Vision Attack Detection for Efficient LVLM Defense |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.19570 |